In a substrate processing apparatus which iteratively performs a substrate process (e.g., a batch process) according to a recipe, monitor data may be obtained at a number of locations or elements (e.g., a temperature sensor, a gas flow meter, a manometer or the like; hereinafter referred to as a “data generation location”). Such monitor data may include, e.g., time-series data including a temperature, a gas flow rate, a pressure or the like, which is representative of the progress of a substrate process or a condition of the substrate processing apparatus. In the related art, a group managing apparatus (i.e., a higher-level management device) has been utilized to collectively and efficiently manage a condition of a group of the substrate processing apparatus or the progress of the substrate process performed therein. Such a group managing apparatus collects monitor data provided from each substrate processing apparatus, which represents the progress of the substrate processes or the condition of the substrate processing apparatus, in a database (DB), and performs a predetermined analysis on the monitor data.
The group managing apparatus retrieves the monitor data stored in the database at a prescribed period (for example, a time interval between the start and end of a process defined in the recipe), produce representative value data (including an average, a maximum, a minimum, a standard variation, etc.) based on the retrieved monitor data, and store the produced representative value data in the database. Further, in the course of storing the currently produced representative value data in the database, the group managing apparatus may also retrieve any previously stored representative value data, which was produced under the same condition. In this case, the group managing apparatus may process the currently produced representative value data and the previously stored representative value data as population parameters and perform a statistical analysis on the population parameters.
However, if even one of improper value data is contained in the representative value data used as the population parameters, the group managing apparatus may generate unexpected analysis results, which leads to erroneous diagnosis of the operational reliability of the substrate processing apparatuses. For example, the group managing apparatus may collect monitor data generated by the substrate processing apparatuses even during a time period that is not subject to the statistical analysis, e.g., when a test run is performed on the substrate processing apparatuses in a production field, or when a maintenance work is performed on the substrate processing apparatus is performed. Further, the group managing apparatus may produce a representative value data (i.e., representative value data that should not be treated as targets for further statistical analysis) based on the collected monitor data and store them in the database as population parameters. In this situation, when statistical analysis is performed on such population parameters, it is difficult to evaluate accurately the operational reliability of the substrate processing apparatuses, which again leads to incorrect diagnosis results. In addition, such representative value data stored in the database may be retrieved for further analysis, which will adversely affect subsequent analysis results.